New car CO2 levels worse than official tests claim

Analysis by consumer group Which? has revealed that the amount of CO₂ produced by new cars is actually increasing, despite official lab tests showing the opposite.

It has long been known that cars use more fuel and produce more pollution in real-world driving than the official lab figures, despite the introduction of more stringent testing in recent years. But the Which? findings suggest that while lab test results are getting better, real-world results are actually getting worse.

Which? does its own emissions testing, which is significantly tougher than the official EU-mandated tests used for working out your road tax and other calculations. This includes a motorway test, where cars are accelerated up to and sustain motorway speeds. In addition, Which? turns on the air-conditioning and adds an extra 200kg of weight (the equivalent of two adults, some luggage and a tank of fuel) to better replicate the way that cars are actually used.

It appears that the manufacturers’ ongoing attempts to game the lab tests are resulting in cars that perform better under very light loads (EU lab conditions) but much worse when worked harder (as driven in real life by real people on real roads), combined with extra equipment and tech designed to reduce emissions using more fuel and therefore actually doing the opposite.

Which? analysed every car it has tested since the start of 2017 – just under 300 different models – and found consistent trends across the board. Small petrol cars (Ford Fiesta, Volkswagen Polo and others) saw an average increase in CO₂ emissions of more than 11% (from 131g/km to 146g/km), while mid-size petrol SUVs (Nissan Qashqai, Kia Sportage and others) rose by 20% (from 158g/km to 190g/km). Worse than either of these categories were large petrol-hybrid cars (like the BMW 3 Series and similar), which recorded an average increase of nearly 32% (from 89g/km to 117g/km).

According to Which?, one of the contributing factors is the ever-increasing weight of new cars. According to its data, the weight of cars tested has increased by an average of 67kg for cars complying with the latest emissions regulations. Some of this increase is additional tech and equipment, but some of it is also additional emissions kit required to reduce pollution coming out of car exhausts.

Table of car emissions (by vehicle type)

Source: Which?

Good news – some toxic pollution is much lower

The Which? analysis did find emissions linked to harming human health – NO_x (oxides of nitrogen) and CO (carbon monoxide) – are now a fraction of what they were, compared to cars made in recent years. In petrol cars, the level of carbon monoxide emissions has dropped by more than 40%, on average, when comparing newer cars tested under the latest regime versus older cars tested under previous regulations. Across all diesel vehicles, the decrease in NO_x was an average of 84%.

These pollutants are directly linked to human health, and have been at the centre of arguments over banning diesel vehicles from urban areas. Diesel engines produce less CO₂ than petrol engines, which is good for global warming, but much more local air pollution in the form of NO_x and CO, which is detrimental to anyone near a diesel exhaust pipe. The reduction in these pollutants is certainly most welcome, although the unfortunate trade-off from improving local air quality is a potential increase in global CO₂ levels.

CO₂ is a big global problem, but NO_x is a big local problem. CO₂ won’t kill you now, but NO_x will. However, CO₂ could kill all of us eventually. This is the trade-off that local and national governments are having to grapple with, and it doesn’t help if they continue to rely on tests that are far less taxing than normal everyday life.

So what needs to happen?

The latest WLTP emissions tests introduced in 2018 were supposed to be more realistic than the antiquated and discredited NEDC tests that were used for decades beforehand (and continue to be used for calculating certain tax obligations, although they are gradually being phased out). Getting the test protocols agreed and legislated was a lengthy painful process as environmentalists pushed for tougher standards while car manufacturers pushed for them to be watered down.

While the new rules are better than the old ones, that’s a bit like saying that an exam score of a C+ is better than an F – it’s still not good enough and it should be better. Obviously it’s a complicated process, but if the testing is less taxing than ordinary everyday use then why bother at all?

From next year, car manufacturers will have to comply with new laws that limit the average CO₂ emissions across their model ranges – based on the flawed lab test, of course. As usual, it’s become complicated and there are loopholes that the industry will seize upon, but the upshot is that we will see many more mild hybrid models (where a small electric motor boosts a petrol or diesel engine, but can’t drive the car on electrical power alone like a full hybrid), more plug-in hybrids and more electric cars.

If the added electrification reduced real-world emissions rather than just helping cars pass lab tests, then that would be a good thing for both local and global emissions. We’ll have to wait and see what future research from Which? and others turns up over the next year or two.